Epigenetic therapy restores normal hematopoiesis in a zebrafish model of NUP98-HOXA9-induced myeloid disease.

Acute myeloid leukemia (AML) occurs when multiple genetic aberrations alter white blood cell development, leading to hyperproliferation and arrest of cell differentiation. Pertinent animal models link in vitro studies with the use of new agents in clinical trials. We generated a transgenic zebrafish expressing human NUP98-HOXA9 (NHA9), a fusion oncogene found in high-risk AML. Embryos developed a preleukemic state with anemia and myeloid cell expansion, and adult fish developed a myeloproliferative neoplasm (MPN). We leveraged this model to show that NHA9 increases the number of hematopoietic stem cells, and that oncogenic function of NHA9 depends on downstream activation of meis1, the PTGS/COX pathway and genome hypermethylation through the DNA methyltransferase, dnmt1. We restored normal hematopoiesis in NHA9 embryos with knockdown of meis1 or dnmt1, as well as pharmacologic treatment with DNA (cytosine-5)-methyltransferase (DNMT) inhibitors or cyclo-oxygenase (COX) inhibitors. DNMT inhibitors reduced genome methylation to near normal levels. Strikingly, we discovered synergy when we combined sub-monotherapeutic doses of a histone deacetylase inhibitor plus either a DNMT inhibitor or COX inhibitor to block the effects of NHA9 on zebrafish blood development. Our work proposes novel drug targets in NHA9-induced myeloid disease, and suggests rational therapies by combining minimal doses of known bioactive compounds.

Characterization of a self-splicing mini-intein and its conversion into autocatalytic N- and C-terminal cleavage elements: facile production of protein building blocks for protein ligation.

The determinants governing the self-catalyzed splicing and cleavage events by a mini-intein of 154 amino acids, derived from the dnaB gene of Synechocystis sp. were investigated. The residues at the splice junctions have a profound effect on splicing and peptide bond cleavage at either the N- or C-terminus of the intein. Mutation of the native Gly residue preceding the intein blocked splicing and cleavage at the N-terminal splice junction, while substitution of the intein C-terminal Asn154 resulted in the modulation of N-terminal cleavage activity. Controlled cleavage at the C-terminal splice junction involving cyclization of Asn154 was achieved by substitution of the intein N-terminal cysteine residue with alanine and mutation of the native C-extein residues. The C-terminal cleavage reaction was found to be pH-dependent, with an optimum between pH6.0 and 7.5. These findings allowed the development of single junction cleavage vectors for the facile production of proteins as well as protein building blocks with complementary reactive groups. A protein sequence was fused to either the N-terminus or C-terminus of the intein, which was fused to a chitin binding domain. The N-terminal cleavage reaction was induced by 2-mercaptoethanesulfonic acid and released the 43kDa maltose binding protein with an active C-terminal thioester. The 58kDa T4 DNA ligase possessing an N-terminal cysteine was generated by a C-terminal cleavage reaction induced by pH and temperature shifts. The intein-generated proteins were joined together through a native peptide bond. This intein-mediated protein ligation approach opens up novel routes in protein engineering.

A topA intein in Pyrococcus furiosus and its relatedness to the r-gyr intein of Methanococcus jannaschii.

A new intein coding sequence was found in a topA (DNA topoisomerase I) gene by cloning and sequencing this gene from the hyperthermophilic Archaeon Pyrococcus furiosus. The predicted Pfu topA intein sequence is 373 amino acids long and located two residues away from the catalytic tyrosine of the topoisomerase. It contains putative intein sequence blocks (C, E, and H) associated with intein endonuclease activity, in addition to intein sequence blocks (A, B, F, and G) that are necessary for protein splicing. This DNA topoisomerase I intein is most related to a reverse gyrase intein from the methanogenic Archaeon Methanococcus jannaschii. These two inteins share 31% amino acid sequence identity and, more importantly, have the same insertion sites in their respective host proteins. It is suggested that these two inteins are homologous inteins present in structurally related, but functionally distinct, proteins, with implications on intein evolution and intein homing.